Although cancer stem cells (CSCs) present an appealing therapeutic target in oral squamous cells carcinomas (OSCCs), a consensus molecular profile for such oral CSCs has not emerged. CSC eradication is further hampered by growing evidence of phenotypic plasticity in some non-CSCs allowing them to return to the CSC pool. Accurately defining oral CSCs together with mechanisms for restoring them after targeted depletion would establish a novel conceptual framework for effective OSCC therapy. Our preliminary work has isolated a drug-resistant subset of slow-cycling JARID1B-hi cells as a candidate CSC population in OSCC. JARID1B, a histone demethylase whose expression identifies CSCs by functional criteria in other cancer types, shows promise as a generalizable CSC marker based on its ability to mediate entry into a stem cell-like epigenetic state. Hyper-activation of the PI3-kinase (PI3K) pathway in JARID1B-hi cells renders them sensitive to depletion by PI3K-targeted drugs. Our working hypothesis is that a depleted JARID1B-hi CSC pool can be replenished by a population of quiescent G0-like cells, which have both high growth potential and drug resistance. While lacking the stem cell markers and PI3K pathway activity seen in JARID1B-hi cells, G0-like cells do show prominent Notch pathway activation, which has been linked to CSC population expansion and increased PI3K/AKT signaling. To explain these findings, our overall hypothesis is that entry of proliferative OSCC cells into a quiescent, G0-like state creates a reservoir to prevent permanent CSC depletion by providing a direct precursor to JARID1B-hi cells. To test this hypothesis, we will first test the roles of JARID1B- hi and G0-like subsets in OSCC growth using conventional CSC criteria (aim 1). We will specifically test whether JARID1B-hi cells have independent CSC functions, while tumor formation by G0-like cells may depend upon the JARID1B-hi fraction. Additional studies will define the precise role of G0-like cells in sustaining the JARID1B-hi population in OSCC (aim 2). Here we test whether Notch-mediated entry of non-CSCs into a quiescent G0-like state forms a direct precursor for JARID1B-hi cells, thus providing a reservoir for CSC repopulation. Validation of these concepts here would establish new model systems to dissect mechanisms underlying the cell state transitions that prevent durable oral CSC depletion during treatment. In doing so, our work will also facilitate translational studies to jointly target two distinct tumor cell phenotypes that cooperatively sustain OSCC growth via the Notch and PI3K pathways.